Content uploaded by Casey Tesfaye
Author content
All content in this area was uploaded by Casey Tesfaye on May 16, 2016
Content may be subject to copyright.
The global survey of physicists: A collaborative effort illuminates the situation of
women in physics
Rachel Ivie, Casey Langer Tesfaye, Roman Czujko, and Raymond Chu
Citation: AIP Conference Proceedings 1517, 53 (2013); doi: 10.1063/1.4794221
View online: http://dx.doi.org/10.1063/1.4794221
View Table of Contents: http://scitation.aip.org/content/aip/proceeding/aipcp/1517?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Women Physicists Speak Again
AIP Conf. Proc. 795, 179 (2005); 10.1063/1.2128321
Summary of the Situation for Women in Physics in Sweden
AIP Conf. Proc. 795, 161 (2005); 10.1063/1.2128313
The Current Situation of Women in Physics in Ireland
AIP Conf. Proc. 795, 133 (2005); 10.1063/1.2128299
Indonesian Women Physicists
AIP Conf. Proc. 795, 131 (2005); 10.1063/1.2128298
The Situation of Women In Physics In Egypt
AIP Conf. Proc. 795, 119 (2005); 10.1063/1.2128292
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
The Global Survey of Physicists:
A Collaborative Effort Illuminates the Situation of
Women in Physicsa
Rachel Ivie, Casey Langer Tesfaye, Roman Czujko, and Raymond Chu
American Institute of Physics
Abstract. The results of the Global Survey of Physicists draw attention to the need to focus on factors other than
representation when discussing the situation of women in physics. Previous studies of women in physics have mostly
focused on t he lack of women in the field. This study goes beyond the obvious short age of women and shows that there
are much deeper issues. For the first time, a multinational study was conducted with approximately 15,000 respondents
from 130 countries, showing that problems for women in physics transcend national borders. Across all countries, women
have fewer resources and opportunities and are more affected by cultural expectations concerning child care. We show
that limited resources and opportunities hurt career progress, and because women have fewer opportunities and resources,
their careers progress more slowly. We also show the disproportionate effects of children on women physicists’ careers.
Cultural expectations about home and family are difficult to change. However, for women to have successful outcomes
and advance in physics, they must have equal access to resources and opportunities.
Keywords: women, physics, international
PACS: 01.75.+m, 01.85.+f, 89.65.Cd
Because women are underrepresented in physics across the much of the globe, discourse about women in physics
centers on this issue of representation. The unspoken assumption seems to be that if the representation of women
were to increase to some higher level, all would be well. However, the focus on representation obscures some
important issues and ignores the day-to-day experiences of women in physics.
In fact, women could be the majority in some hypothetical world of physics yet still experience problems in their
careers. These problems could stem from bias, often unconscious, against women in science. But do women actually
experience problems in their day-to-day work as physicists? Do they have equal access to opportunities and
resources needed to do their work? If not, how does this inequity affect their careers? Furthermore, many cultures
assign the responsibility of child care and housework to women. Is this true for female physicists, and if it is, do
household and family responsibilities affect their careers? If sex-based differences such as these exist, problems for
women in physics will require solutions that encompass more than just increasing their representation.
HISTORY OF THE GLOBAL SURVEYS OF PHYSICISTS
The global surveys of physicists were originally designed to answer just these questions. What are women’s
experiences in physics? What has contributed to their successes, and likewise, which obstacles have they faced? To
answer these questions, the first and second International Conference of Women in Physics sponsored surveys that
were designed to document the situation of women in physics. These conferences were sponsored by the Working
Group for Women in Physics of the International Union of Pure and Applied Physics (IUPAP). The situation as
measured by the surveys encompassed far more than the representation of women in physics in various countries.
The surveys were designed to measure working relationships in graduate school and beyond, access to opportunities
and resources, and the effects of family obligations on careers.
The first two global surveys of physicists were based on the notion that the situations of women in physics
should be documented in order to ascertain how women experience their careers and educational pathways. The goal
was to describe common problems that women in physics across the world face in their work and studies. The first
aSupported by the Henry Luce Foundation and the National Science Foundation, Award 1012148, Beverly Hartline, Principal Investigator.
Women in Physics
AIP Conf. Proc. 1517, 53-61 (2013); doi: 10.1063/1.4794221
© 2013 American Institute of Physics 978-0-7354-1140-1/$30.00
53
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
two global surveys, therefore, were sent exclusively to female physicists. More than 1,000 women from more than
50 countries responded to each survey.
For the third International Conference of Women in Physics, held in Seoul, South Korea, in 2008, the IUPAP
Working Group for Women in Physics decided to expand the scope of the surveys. First, by limiting the survey
respondents to only female physicists, we were missing an essential opportunity to document the differences
between men’s and women’s experiences. Therefore, men were included in the third survey, which took on the title
“The Global Survey of Physicists.” In addition, the IUPAP working group decided to expand the languages of the
survey because the first two surveys had been conducted in English only. With support from the Henry Luce
Foundation, the survey was translated into seven languages other than English. These included all the UN languages
(Russian, Arabic, Spanish, Chinese, and French), in addition to Japanese and German.
The Statistical Research Center (SRC) of the American Institute of Physics (AIP), which had conducted the first
two surveys, again undertook this project and began by working collaboratively with women’s working group team
leaders from the IUPAP countries. These team leaders were called on to contribute surveys that they had conducted
or knew of in their own countries. The SRC analyzed the content of these questionnaires so that the newly revised
third survey could include issues that were relevant and common across all contributed surveys. The survey was
reworked to accommodate the new content, submitted to the women’s working group team leaders for feedback, and
translated. We then called on the women’s working group team leaders to comment on the translated versions of the
surveys to ensure their viability in the field. Involving the team leaders in more aspects of the process both enhanced
the collaborative nature of the project and dampened any natural propensity to approach a multicultural project from
a limited cultural perspective.
Once the surveys were ready to be distributed, most of the team leaders distributed the web-based survey among
their contacts. At the end of the survey form, respondents were encouraged to pass the survey on to other physicists,
especially women, whom they knew. This created a snowball distribution. In some countries, physical societies
distributed the survey to their members. For example, the American Physical Society and the German Physical
Society distributed the survey to random samples of their members, while the Japanese Physical Society distributed
the survey to all its members. The results of our analysis in this article will be based on the entire group of
respondents from all countries.
The survey was available to respondents for one year, from October 2009 to October 2010. At the close of the
survey, 14,932 physicists from 130 countries had responded, a dramatic increase from the 1,000+ who had
responded to past surveys. Part of the reason for the huge increase in respondents was the addition of men. But the
distribution by physical societies, the addition of seven languages, and the extra help and involvement from the team
leaders no doubt contributed to the increase as well. In the end, 22% of the respondents were women, representing
approximately 3,000 women, a significant increase over the number answering the first two women-only surveys.
Three-quarters of the respondents to the survey were from countries that the UN defined as “very highly
developed.” In this analysis, we will distinguish between respondents from the UN-defined very highly developed
countries [1] and respondents from less developed countries. The analyses in this paper are limited to respondents
who are not students and who have had a job that uses their knowledge of or skills in physics.
IMPORTANCE OF RESOURCES, OPPORTUNITIES,
AND FAMILY RESPONSIBILITIES
Without access to resources and opportunities needed to conduct research and disseminate results, careers in
science can stall. There are many types of resources needed, ranging from access to graduate students or employees
to assist with research, to clerical support, research funding, and travel money. However, whether or not these
resources are distributed equitably to women and men in physics has not been documented. One study of social
science faculty members in the United States found that mothers are less likely than fathers and childless professors
to have access to resources, but this difference was entirely explained by mothers being more likely to work outside
of research-intensive universities [2]. Although it would not make sense, given the different educational systems
across the globe, to compare types of universities, we will examine whether women have access to just as many
resources as men and whether these differences can be explained by differences in their type of employer. We will
also consider the role of professional opportunities, which are essential to career advancement for scientists.
Opportunities that can advance scientists’ careers include invitations to speak, serving on committees, and
conducting resources abroad. As with resources, we ask, are these opportunities equally available to women and
men?
54
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
The effects of cultural expectations requiring women to take most of the child care and household
responsibilities cannot be overlooked. These effects have been documented in the U.S. for academic women’s career
outcomes. One of the most cited studies found, among many other things, that mothers in the U.S. are 29% less
likely to enter tenure-track positions than women without children. Furthermore, “successful” academic women, that
is, women who are full professors, are much less likely to be married with children than male full professors [3].
Clearly cultural expectations vis-à-vis marriage and family affect the broad class of U.S. female faculty members.
Will the link between family obligations and careers hold for physicists, and will it hold in other countries?
Limited Resources
We asked respondents whether they had enough of the following items to do their research: funding, office
space, lab space, equipment, travel money, clerical support, and employees or students. The percentages of
respondents who said yes are given in Table 1. Even in this simple analysis, when comparing the answers for
women and men in very highly developed countries versus other countries, we can see that women are less likely to
have access to these resources than men are.
The women who answered the survey tended to be younger than the men. Could this account for the difference
in resources, because younger people may be expected to have fewer resources? Could different types of jobs
explain the difference in resources between men and women as it did in the social science faculty study described
above [2]? In order to determine if the sex difference can be accounted for by these factors, we ran binomial logistic
regressions on each of the items, including independent variables for age of respon dents, sector of employment
(university, private sector, government, and other), and country’s level of development. This multivariate analysis
showed that even when controlling for these other factors, the sex difference was still significant. In other words,
regardless of age, country’s level of development, and even across all employment sectors, women reported having
access to fewer resources than men (Table 1).
TABLE 1. Percentage of Responding Physicists Who Are Currently
Employed and Said They Had Enough Resources to Conduct or
Present Their Research
Women
Men
Women
Men
Funding
34
51
52
60
Office space
64
74
72
77
Lab space
42
47
46
52
Equipment
42
49
58
64
Travel money
31
47
57
64
Clerical support
22
38
30
43
Employees or students
42
53
33
43
N=
384
1,187
1,140
4,947
TABLE 2. Percentage of Responding Physicists Who Said They Had Career Opportunities
Less Developed
Very Highly Developed
Women
Men
Women
Men
Given a talk at a conference as an invited speaker*
51
67
58
73
Advised graduate students*
63
77
58
70
Conducted research abroad*
54
71
61
69
Acted as boss or manager*
38
53
46
61
Served as editor of a journal*
16
24
11
19
Served on committees for grant agencies*
22
37
26
36
Served on thesis or dissertation committees (not as an advisor)*
52
66
37
52
Served on important committees at your institute or company
50
62
48
60
Served on an organizing committee for a conference in your field
48
59
48
55
Advised undergraduate students
82
84
69
74
Attended a conference abroad
75
81
83
87
N=
405
1,209
1,234
5,198
*The asterisks indicate items that men were more likely to do than women (p<.05). Women were just as likely to do the remaining four
items as men were. There were no items that women were more likely to do than men.
55
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
Lack of Professional Opportunities
As we stated at the beginning of this article, a focus on women’s representation often overlooks other potential
sources of inequality. One of these includes professi onal opportunities. We asked our respondents whether they had
participated in a variety of experiences, many of which would serve to advance almost any scientist’s career
(Table 2). An examination of these percentages shows that there is a sex difference in most of these opportunities,
with women being less likely to have these experiences than men. To make sure that the sex difference still exists
across different ages, sectors of employment, and country’s level of development, we again ran binomial logistic
regression on these items. We found that for most of the items, the sex difference was robust; women still had fewer
opportunities than men, even accounting for age, type of employment, and country’s development. However, four of
the items—advising undergraduates, attending conferences abroad, and two types of committee work—no longer
had statistically significant sex differences. Variations on these items were better explained by age and type of job
than by differences between men and women. There were no items that women were more likely to do than men.
Family Responsibilities
In much of the world, women are the ones who hold primary responsibility for taking care of the home and
children. Studies of the careers of faculty members [3] and the first two global surveys of women physicists
repeatedly and consistently document the effects of children on women’s careers [4, 5]. Given that foundation, we
decided to turn again to the effects of children and home responsibilities on physics careers, but this time we
examined whether family responsibilities affect women’s and men’s careers differently.
First, we note that female physicists in this study are more likely than men to have a partner or spouse with a
high level of education (Figure 1). Men are more likely to be married or partnered with someone who either does not
work or who earns less money than they do (Figure 2). We also note that when asked directly about which partner is
responsible for the majority of housework, women report doing more of the housework than men, and men are more
likely to report that their spouses are responsible for most of the housework (Figure 3). What do these findings mean
for the careers of men and women in physics?
Taken together, the results on these three charts indicate that if family responsibilities do affect physicists’
careers, they will be more likely to do so for women than for men. Not only are men less likely to do housework,
they are more likely to have spouses or partners who are available to do it for them. When sacrifice needs to be
made in order to care for a sick child or family member, it makes economic sense for the partner who makes less
money to take on that responsibility. And for most men, the partner who makes less money is someone else, thus
enabling the men to focus on their careers without these distractions. In addition, recall that the women in this study
are working physicists. These results show that even when the women are working, they are still more likely to be
responsible for household duties.
FIGURE 1. Spouse’s highest level of education.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Women Men Women Men
Less Developed Very Highly Developed
Doctoral degree
Some doctoral study
University
Some university
Primary or secondary
56
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
FIGURE 2. Employment and earning status of spouses.
FIGURE 3. Responsibility for majority of housework, respondents with employed spouses only.
Career Progress
Effect of family responsibilities on career progress. Based on the sex differences that we have seen in
housework and family responsibilities, we hypothesize that female respondents will report slower career progress
than men. First, we will examine several measures of career progress, taking into account the division of family
responsibilities. In the survey, we asked respondents a specific set of questions about how their work or careers
changed when they became parents. The results in Table 3 show that women are much more likely to report that
becoming parents affected their work than men are. Notably, the women in our study were about twice as likely to
report that they chose different schedules and spent less time at work. Women also became more efficient at their
work. These findings echo the results from the first two global surveys of physicists, in which women reported that
one major effect of having children was that they became more efficient due to having to leave the lab or office
earlier to pick up a child from their child care provider , for example. More often than men, women reported that
their careers or rates of promotion slowed significantly after becoming parents. All of the differences shown in
Table 3 are significant in the multivariate logistic regression models we tested. As before, we controlled for age,
sector of employment, and country’s level of development. Even differences that appear small in Table 3 are
significant in these multivariate models, meaning that regardless of age, type of employment, or country’s
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Women Men Women Men
Less Developed Very Highly Developed
More money than you
About the same amount
Less money than you
Spouse is not employed
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Women Men Women Men
Less Developed Very Highly Developed
Employed domestic workers
Shared equally by the family
members
Other family members
My partner or spouse
I do most of the
housekeeping
57
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
development, women reported altering their behavior after becoming parents more often than men did. In fact, as the
last item in the table indicates, men were much more likely than women to say that that their careers did not change
much at all when they became parents.
Next, we turn to respondents’ perceptions of their overall career progress. Our previous studies [4, 5] showed
that women with children were more likely to say that their careers had progressed more slowly than colleagues who
completed their final degrees at the same time. The third Global Survey of Physicists allows us to compare this
effect by sex to determine the effect of children on all physicists’ career progress. The results show that women with
children are the most likely group to say that their careers had progressed more slowly than their colleagues
(Figure 4).
FIGURE 4. Career progress by sex and parenthood.
For this question, we again tested a multivariate model using ordinal logistic regression. In addition to the
previously used independent variables (age, employment sector, and country’s level of development), we included a
variable for having a child and also tested for an interaction effect between having a child and sex to determine if
parenthood affects men’s and women’s careers differently. These results show that women with children are more
likely than women without children and men to say that their careers had progressed more slowly, even controlling
for age, sector of employment, and country’s development. In fact, as demonstrated in Figure 4, men with children
are the least likely to say their careers had progressed slowly.
We have demonstrated the effect of children on career progress, but it is also likely that career choices
themselves cause people to think and act differently when making decisions about their family life. Our previous
studies [4, 5] suggest that some women, particularly older women, deliberately chose not to marry or not to have
children in order to focus on physics. So for this third survey, we asked respondents about the effects of their careers
on decisions about marriage and children. Women were more likely than men to say that their careers affected their
decisions about marriage and family (Figure 5). We again tested this effect in a multivariate model and found that
controlling for age, sector of employment, and country’s level of development, female physicists who responded
were more likely than male physicists to say that their careers affected their family lives.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Children No Children Children No Children
Women Men
More quickly
About the same
More slowly
TABLE 3. Percentage of Responding Physicists Who Reported Changes When
Becoming Parents
Women
Men
My work or career did not change significantly
32
65
I chose a less demanding or more flexible work schedule
39
20
I spent significantly less time at work
35
18
I was more productive and efficient at work
29
15
My career or rate of promotion slowed significantly
34
9
I changed my employer or field of employment
7
4
I became a stay-at-home parent
6
1
58
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
FIGURE 5. Respondents saying that careers affected family life.
Career progress, resources, and opportunities. Being a mother does affect career advancement for the
respondents in this survey. We have also shown that there are sex differences in resources and opportunities that
could advance physicists’ careers. Our final step is to document the empirical effects of resources and opportunities
on career advancement.
Figures 6 and 7 present totals for measuring resources and opportunities, with the top row representing
respondents who had only one resource or one opportunity, an d the bottom r ow representing respondents who said
they had them all. The data in Figures 6 and 7 demonstrate that the more resources and opportunities respondents
had, the more likely they were to say that their careers progressed more quickly than their colleagues. We also tested
this relationship in multivariate models, which showed that controlling for sex, country’s level of development, age,
and sector of employment, resources and opportunities do affect career advancement. One interpretation of this
finding is that for both men and women, having resources and opportunities serves to advance respondents’ careers
more quickly.
Our measure of career advancement is subjective, but the results are consistent with studies of objective
measures of advancement, such as receiving tenure, that have been conducted in the U.S. [3]. In this study we found
it necessary to use subjective measures because these measures work across all types of employment sectors and
educational systems.
Could sex differences in resources, opportunities, and career advancement exist, not because women are treated
differently, but because they have different expectations than men? Although our measures are subjective and such
an explanation is possible, the fact that our results are consistent with previous research suggests that women’s
experiences, not their expectations, are different.
FIGURE 6. Effect of opportunities on career advancement for male and female physicists.
0
20
40
60
80
100
Women Men Women Men
Less Developed Very Highly Developed
Percent Yes
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
1
2
3
4
5
6
7
8
9
10
11
Opportunities Total
More quickly About the same More slowly
59
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
FIGURE 7. Effect of resources on career advancement for male and female physicists .
CONCLUSION
The Global Survey of Physicists follows a body of work that examines the importance of access to resources and
opportunities on success in careers. We found that female respondents were less likely than male respondents to
report access to a variety of resources that would be helpful in advancing a scientific career. Women were also less
likely to report having access to professional opportunities and positions overall than men who responded.
Given the unequal distribution of resources and opportunities, and the cultural expectations that women will be
primarily responsible for the household and child care, we expected to find that women’s career progress in physics
would be slower than that of men’s progress. Indeed, this is what we found on multiple measures of career progress.
Women are more likely than men to have changed their work situations upon becoming parents. Mothers are more
likely than all men and women without children to report that their careers have progressed more slowly than
colleagues who finished their degrees at the same time. Furthermore, women are more likely than men to report that
their careers affected the decisions they made about marriage and children. All these differences again hold true
across age, employment sector, and country’s level of development. The results are summarized in Figure 8.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
1
2
3
4
5
6
7
Total Resources
More quickly About the same More slowly
Opportunities
and resources
Employment
sector
Country’s human
development
index
Sex
Age Sex*Children
Career
progress
FIGURE 8. A model of career advancement for physicists.
60
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
These results will come as no surprise to researchers who have already found effects of resources, opportunities,
and family responsibilities on women’s careers [2, 3]. The results are also consistent with the sociological concept of
cumulative disadvantage [6], which posits that disadvantages (or advantages) lead to subsequent disadvantages (or
advantages) in almost a chain reaction of disadvantage (or advantage). The disadvantages that women experience in
resources and opportunities, along with the expectation of devoting more time to home and family, have the
cumulative effect of slowing down women’s physics careers in comparison to men’s.
This article documents the limited resources and professional opportunities that female physicists in our study
experienced when compared to male physicists. It also supports the hypothesis that these disadvantages and the
effects of family expectations affect women’s career outcomes in physics. Although physics continues to have a low
representation of women compared to other fields, this article documents the need to address problems beyond the
low numbers of women in physics. We must also address the inequities in resources and opportunities that limit
women’s progress in physics careers. For women to have successful outcomes and advance in physics, they must
have equal access to resources and opportunities. Cultural expectations about home and family are much more
difficult to change, but inequities in opportunities and resources can be remedied. Resources are of course limited
across any economic system, but it is unjust to distribute them, even unconsciously, according to sex, a fixed
characteristic over which individuals have no control. The science of the future should be science for all.
REFERENCES
1. UN Human Development Index, accessed April 17, 2012, http://hdr.undp.org/en/statistics/.
2. R. Spalter-Roth and W. Erskine, Resources or Rewards: The Distribution of Work-Family Policies, 2006, accessed April 17,
2012, www.asanet.org/images/research/docs/pdf/Resources%20or%20Rewards.pdf.
3. M.A. Mason and M. Goulden, “Marriage and Baby Blues: Re-defining Gender Equity,” 2004, accessed April 17, 2012,
http://ucfamilyedge.berkeley.edu/marriagebabyblues.pdf.
4. R. Ivie and S. Guo, Women Physicists Speak Again, 2006, accessed June 15, 2012, www.aip.org/statistics/trends/
reports/iupap05.pdf.
5. R. Ivie, R. Czujko, and K. Stowe, Women Physicists Speak, 2002, accessed June 15, 2012, www.aip.org/statistics/trends/
reports/iupap.pdf.
6. G.J. Johnson and W.R. Johnson, “Cumulative Disadvantage Theory and Contingent Work: Race and Gender Comparisons,”
2005, accessed June 15, 2012, www.rcgd.isr.umich.edu/prba/perspectives/fall2005/jones.pdf.
61
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions IP: 198.23.202.152 On: Thu, 28 Apr 2016 20:36:52
